How are founders making product demos, ads, and social video with AI (Veo, Sora, Runway, HeyGen, ElevenLabs)?

The playbook that emerged in 2025: write the script yourself, use Gemini or ChatGPT to turn it into a shot-by-shot prompt list, generate clips in Veo (inside Google Flow), Sora 2, or Runway, then cut it together in CapCut or Premiere with an ElevenLabs voiceover. The reference case is Kalshi's NBA Finals ad, made by one AI filmmaker in three days for about $2,000. For SaaS product demos specifically, founders skip generation entirely and pair real screen captures with AI demo tools like Supademo, since showing the actual product beats synthesizing it.

A quick orientation. The real value is below: resources worth your time, from people who've actually done it.